What is Machine Learning?

What is Machine Learning?

Types of Machine Learning – Supervised Learning

Supervised Learning: - Uses labeled data to train models. - Example: Spam detection in emails (spam vs. non-spam). - Common algorithms: Linear Regression, Decision Trees, Support Vector Machines (SVM), Neural Networks.

Types of Machine Learning – Unsupervised Learnin

Unsupervised Learning: - Finds patterns in unlabeled data. - Example: Customer segmentation in marketing. - Common algorithms: K-Means Clustering, Hierarchical Clustering, Principal Component Analysis (PCA).

Types of Machine Learning – Reinforcement Learning

Reinforcement Learning: - Optimizes decision-making through rewards. - Example: Training an AI to play a game like Chess or Go. - Key components: Agent, Environment, Reward Signal.

Key Components of an ML Model

• Data:
  • The quality and quantity of data are fundamental.
  • Data preprocessing (cleaning, normalization, feature extraction) is crucial.
• Model:
  • A mathematical representation of the problem.
  • Chosen based on the problem type (classification, regression, clustering).

Key Components of an ML Model

• Error function:
  • Evaluates the difference between prediction and actual value.
  • Example: Mean Squared Error (MSE) for regression, Cross-Entropy Loss for classification.
• Optimization:
  • Algorithms that adjust the model parameters to minimize error.
  • Common optimization techniques: Gradient Descent, Adam Optimizer.

Bias and Inductivity

• Inductive Bias:
  • Prior assumptions that the model uses to generalize.
  • Example: Linear models assume data relationships are linear.
• Sample Bias:
  • Differences between training data and real-world data.
  • Example: A face recognition system trained on a specific demographic may perform poorly on others.

Bias and Inductivity

• Bias-Variance Tradeoff:
  • High Bias (Underfitting): The model is too simple, failing to capture patterns.
  • High Variance (Overfitting): The model memorizes training data but fails on new data.

Example of Bias and Variance

Basic Machine Learning Algorithms

1. Linear Regression (Supervised Learning - Regression)

• Predicts a continuous value based on input features.
• Equation: ( y = mx + b )
• Example: Predicting house prices based on square footage.

LinearRegression()

LinearRegression()

Slope: [0.6]

Intercept: 2.2

Basic Machine Learning Algorithms

2. Decision Trees (Supervised Learning - Classification & Regression)

• Splits data into decision nodes to make predictions.
• Example: Diagnosing a disease based on symptoms.

DecisionTreeClassifier()

DecisionTreeClassifier()

Prediction for [1,1]: [1]

Basic Machine Learning Algorithms

3. K-Means Clustering (Unsupervised Learning)

• Groups similar data points together.
• Example: Customer segmentation in marketing.

Cluster Centers: [[10.  2.]
 [ 1.  2.]]

Basic Machine Learning Algorithms

4. Support Vector Machines (SVM) (Supervised Learning - Classification)

• Finds a hyperplane that best separates different classes.
• Example: Classifying tumors as benign or malignant.

SVC()

SVC()

Prediction for [1,1]: [1]

Basic Machine Learning Algorithms

5. Reinforcement Learning Example

• Uses rewards and penalties to train an agent to make optimal decisions.
• Example: A robot learning to navigate a maze.

Trained Q-Table:
 [[0.22876493 0.25104384 0.32798898 0.44926391 0.48521205]
 [0.30181891 0.33454514 0.42135597 0.5727343  0.76654889]
 [0.39806006 0.56751866 0.69480353 0.98485257 1.22712499]
 [0.49995044 0.95288361 1.7589996  2.11821827 3.48281241]
 [0.63288801 1.00579933 2.84112154 5.90826252 0.        ]]